Google on Tuesday, May 20, announced Veo 3, its latest artificial intelligence (AI) video generation model, which now includes the ability to generate audio alongside visuals, a feature that sets it apart from rivals like OpenAI’s Sora.
The new model can generate dialogue between characters, animal sounds, and other forms of synchronised audio, in addition to rendering realistic physics and accurate lip-syncing, according to the company, said a report by CNBC.
“Veo 3 excels from text and image prompting to real-world physics and accurate lip syncing,” said Eli Collins, VP of product at Google DeepMind, in a blog post.
Veo 3 is available starting Tuesday in the US via Google’s new $249.99 per month Ultra subscription plan, aimed at AI power users. It is also accessible to enterprise customers through Google’s Vertex AI platform.
Alongside Veo 3, Google introduced Imagen 4, the latest version of its image generation tool, which the company claims delivers improved quality and responsiveness to prompts, added CNBC.
Google also announced Flow, a new AI filmmaking assistant that lets users describe shots, locations, and visual styles to generate cinematic videos. Flow will be accessible through platforms including Gemini, Whisk, Workspace, and Vertex AI.
These updates come as generative video and image tools see rising interest among both creators and developers. In March, OpenAI CEO Sam Altman noted that demand for ChatGPT-4o’s image generation capabilities briefly overwhelmed its hardware infrastructure.
